SUMMARY Normal splicing of pre-mRNAs is a major source of eukaryotic transcript diversity, and defective pre-mRNA splicing has emerged as a common hallmark of cancers and myeloid neoplasms. The goal of this proposal, ?Molecular Recognition in Pre-mRNA Splicing? is to understand regulation of the early stage of spliceosome assembly, which is commonly affected among human diseases. In so doing, we seek to identify vulnerabilities as potential therapeutic targets. Here, we focus on essential U2AF65 and SF3B1 splicing factors, which together recruit U2 small nuclear ribonucleoproteins to the 3 splice sites of pre-mRNAs. In the prior funding period, we revealed that U2AF65 recognizes a nine-nucleotide polypyrimidine tract splice signal. We mapped cancer-associated mutations on our U2AF65 structures and found clusters of U2AF65 hotspots at key interfaces. In Aim 1 of the coming period, we will (1A) complete our views of U2AF65 recognizing different splice site signals and (1B) test the hypothesized effects of cancer-associated U2AF65 mutations on 3 splice site signal recognition. The results of Aim 1 will aid algorithms to predict splice sites and the impacts of inherited splice site mutations, and importantly establish the molecular consequences of acquired U2AF65 mutations in cancer. In the prior funding period, we also showed that a U2AF65 paralogue, CAPER?, associates with the SF3B1 subunit both in human cell lysates and with purified proteins. In Aim 2 of the coming period, we will (2A) view the SF3B1? U2AF65 interface and test its hypothesized relevance for pre-mRNA splicing in human cells. Then we will (2B) investigate hypothesized phosphorylation-dependent regulation of SF3B1 associations with U2AF65, CAPER?, and other alternative splicing factors during the splicing pathway. In support of the feasibility of our aims, we have prepared diffracting crystals for structure determinations, methods for splicing factor co- immunoprecipitation, knockdown, and re-expression, and we have identified U2AF65- and SF3B1-sensitive splicing substrates as a basis to test our structure-guided hypotheses. SF3B1 is the most frequently mutated splicing factor in human malignancies. Although SF3B1 hotspots are distinct from the U2AF65- and CAPER?- binding sites, a ?double whammy hit? that interferes with the functions of these collaborating factors, e.g. kinase inhibitors such as already have been approved for cancer treatment, would selectively kill cancer cells carrying SF3B1 mutations. Therefore it is important to understand the role of U2AF65 and the influence of phosphorylation for SF3B1 functions. Altogether, the results of these aims will elucidate molecular mechanisms of 3 splice site recognition, and lay a foundation for future therapeutic strategies to treat cancers carrying acquired splicing factor mutations.